
Abiogeography-basedoptimizationalgorithmwithmutationstrategiesformodelparameterestimationofsolarandfuel

cells
QunNiua, ,LetianZhanga,KangLib
ab
SchoolofMechatronicEngineeringandAutomation,ShanghaiKeyLaboratoryofPowerStationAutomationTechnology,ShanghaiUniversity,Shanghai200072,ChinaEnergy,PowerandIntelligentControl,SchoolofElectronics,ElectricalEngineeringandComputerScience,Queen’sUniversityBelfast,BelfastBT95AH,UK
articleinfoabstract
Mathematicalmodelsareusefultoolsforsimulation,evaluation,optimaloperationandcontrolofsolarcellsandprotonexchangemembranefuelcells(PEMFCs).Toidentifythemodelparametersofthesetwotypeofcellsef ciently,abiogeography-basedoptimizationalgorithmwithmutationstrategies(BBO-M)isproposed.TheBBO-Musesthestructureofbiogeography-basedoptimizationalgorithm(BBO),andboththemutationmotivatedfromthedifferentialevolution(DE)algorithmandthechaostheoryareincorpo-ratedintotheBBOstructureforimprovingtheglobalsearchingcapabilityofthealgorithm.Numericalexperimentshavebeenconductedontenbenchmarkfunctionswith50dimensions,andtheresultsshowthatBBO-Mcanproducesolutionsofhighqualityandhasfastconvergencerate.Then,theproposedBBO-Misappliedtothemodelparameterestimationofthetwotypeofcells.TheexperimentalresultsclearlydemonstratethepoweroftheproposedBBO-Minestimatingmodelparametersofbothsolarandfuelcells.
Ó2014ElsevierLtd.Allrightsreserved.
Articlehistory:
Received3January2014Accepted10June2014
Keywords:SolarcellPEMfuelcell
Biogeography-basedoptimizationMutationstrategyParameterestimation
1.Introduction
Thegreenhouseemissionsandfossilfuelsshortagearethetwinchallengesfacingthemankind,andthedevelopmentofnewcleanenergytechnologieshasbeenrecognizedasthekeytotackletheproblems.Solarcellsandprotonexchangemembranefuelcells(PEMFCs)haveattractedconsiderableattentionduetothefeaturesoftheirhighpowerdensity,nopollutionandlownoise.Asphoto-electricconverters,solarcellsnotonlyhavenobyproductbutalsohavetheadvantagesofeasyinstallationandlittlemaintenancecost.Similarly,PEMFCsgenerateelectricitybyoxidizinghydrogenunderrelativelowoperatingtemperatureswithnon-pollutingbyproducts,andhavebeenwidelyusedinvehiclepropulsion,smalldistributedgenerationandportableapplications[1–4].
ModelingtheoperationofsolarcellsandPEMFCshasbecomeasigni cantresearchtopicinrecentyearsinordertobetterexploreandutilizetheirproperties.Somemodelshavebeenproposedtosimulatethebehaviorsofbothcellsunderdifferentoperatingcon-ditions[5–10],andmostofthemarelessusefulinpracticalappli-cationsduetothecomplexityandproblemspeci c.Inrecentyears,mathematicalmodelswithsomeunknownparametersarewidelyusedinpractice[11–13].Theseparameterswhichre ecttheactualperformances(suchasairpressure,fuel owrate,celltemperature,Correspondingauthor.Tel.:+8602156334241;fax:+8602156334241.
E-mailaddress:comelycc@hotmail.com(Q.Niu).

doc.guandang.net.
diffusionandrecombinationdioderesistance)areprimarilychosenbasedonexperience,andtheselow-accurateempiricalvaluesinev-itablyaffecteffectivenessofthedevelopedmodelsinsimulation,design,optimaloperationandcontrol.Therefore,itisindispensabletoinvestigatetheproblemsofparameterestimationproblemsofthesolarcellandPEMFCmodels.
Someconventionaldeterministicoptimizationmethodshavebeenusedtosolvetheparameterestimations[14–18].However,thesetwokindsofcellsarenonlinear,non-convexandstronglycou-pledsystemsandmostoftheseconventionalapproachesrequirecontinuity,convexityanddifferentiabilityconditions,whichmakethemhardtohandletheparameterestimationproblemsmentionedabove.Withthedevelopmentofcomputingtechnologiesandarti -cialintelligence,manymeta-heuristicoptimizationalgorithms,whichareofconceptualandcomputationalsimplicity,beingexcel-lentreal-worldproblemsolversandrobusttodynamicenviron-ments,capableofsolvingproblemswithnoknownsolutionsandwithnoneedforanalyticexpressionoftheproblems[19],havebeenappliedtotheparameterestimationproblemsofsolarcellandPEM-FC.Geneticalgorithm(GA)[20–23],simulatedannealing(SA)[24,25],particleswarmoptimization(PSO)[26–29],differentialevolution(DE)[30,31]andarti cialneuralnetworks(ANNs)[32,33]havebeenproposedtoimprovetheparameteraccuracyforbothcellmodelsrespectively.Besides,severalothermethodsincludingbirdmatingoptimizer(BMO)[34,35]andseekeroptimizationalgorithm(SOA)[36]havealsobeenusedtosolve
the DE algorithm employs simple mutation and crossover operators to generate...optimization, in which both trial vector generation strategies and the ...
Hybrid Algorithm, Multimo? dal,Optimization. ...computing strategies, such as gradient, have ...with 1/3 and the mutation operators that result...
evolution algorithm with ensemble of parameters and mutation strategies R. ...c optimization problem, it is generally necessary to perform a time-...
biogeography-based optimization (CBBO) algorithm for...Mutation tends to take place on habitats with ... we can adopt several new strategies to avoid ...
Additionally, the mutation mechanism also used to ...optimization, is a population-based algorithm and ...Region growing strategies start at known pixels ...
Biogeography-based optimization algorithm (BBO) is a relatively new ...Table 4 shows these procedures: 3.2.2 mutation model with mutation ...
multimodal and deceptive benchmark functions when used with low mutation ...probabilistic analysis of a simpli?ed biogeography-based optimization algorithm....
The existing crossover strategies in DE show ...mutation based search method in genetic algorithm ...“Ant Colony Optimization with increasing population...
optimization algorithm (HIMO) based on clonal ... a hybrid mutation operator is proposed with the...Genetic search strategies in multicriterion optimal ...
Vol.15, No.2 An Ant Colony Optimization Algorithm Based on Mutation and Dynamic Pheromone U_专业资料。Zhu QB, Yang ZJ. An ant colony optimization ...
热门文档